19--Spamalot--Kathleen Rauh

[nckathleen]

• [x] Blank slate
  • [x] Create a GitHub repo called spambase
  • [x] Set up your requirements/virtual environment
  • [x] Download the spambase dataset, but do not commit it to your repository
  • [x] Create an IPython notebook to write your code and collect your findings
• [x] Normal mode
  • [x] Load in the data file, doing any cleaning necessary to get usable data
  • [x] Subsample the data set into training and test data
  • [x] Write code to classify the data into spam/not-spam, making sure that you just use your training data to build your model, and checking your results with your test data
• [ ] Hard mode
  • [x] Try reducing or changing your features in order to get better results
  • [ ] Find another dataset and break it down into features
  • [ ] Test your algorthm on the new dataset and compare its performance

[nckathleen]

https://github.com/nckathleen/Spambase Did some hard mode, too. What!?

@jamesmallen @powder-river

[powder-river]

@nckathleen

WHAT!!! (MIKE DROPPED)

awesome hw! way to get in a hard mode attempt too!